DoorDash-Specific ML Applications Questions
Domain-specific machine learning use cases within the DoorDash platform, covering production ML lifecycle topics such as demand forecasting, driver dispatch and routing, pricing and revenue optimization, recommendations, fraud detection, and real-time optimization. Includes model development, deployment, monitoring, drift handling, and scalability considerations for ML systems in a high-velocity delivery marketplace.
HardTechnical
0 practiced
Write Python pseudocode to detect when a driver becomes inactive given a stream of GPS pings (driver_id, lat, lon, ts). Account for GPS jitter, varying ping intervals, expected short breaks, and noisy signals. Describe tuning parameters, complexity, and how to reduce false positives.
EasyTechnical
0 practiced
Implement a baseline forecasting function in Python that predicts next-hour order volume for a zone using the previous 24 hourly counts. Input: a list of 24 hourly integers (most recent last) with possible nulls for missing hours. Return a single numeric forecast for the next hour and describe complexity and assumptions in comments.
HardTechnical
0 practiced
Compare privacy-preserving ML techniques relevant to DoorDash such as federated learning, secure aggregation, and differential privacy. Discuss the utility-privacy trade-offs, engineering and deployment complexity, and regulatory or compliance considerations when collecting customer and driver behavioral signals across regions.
HardTechnical
0 practiced
Implement a Python function that estimates expected driver pickup distance for a given order location using Monte Carlo sampling over a provided list of driver (lat, lon) locations and a simple acceptance probability model. Inputs: drivers list, order location, N samples; output: expected distance in meters. Discuss optimizations for serving many orders in batch.
HardTechnical
0 practiced
Propose an identification strategy to estimate the causal effect of price discounts on short-term order volume and long-term customer retention at DoorDash. Compare randomized controlled trials with observational methods (instrumental variables, difference-in-differences, synthetic controls), and discuss threats to identification like spillovers and selection bias.
Unlock Full Question Bank
Get access to hundreds of DoorDash-Specific ML Applications interview questions and detailed answers.
Sign in to ContinueJoin thousands of developers preparing for their dream job.